TL;DR
This paper introduces a novel three-stage Data Envelopment Analysis (DEA) approach to accurately assess the operational efficiency of China's air transport sector across different provinces, accounting for external economic and social factors. The findings reveal significant regional disparities in efficiency and offer tailored strategies for improvement, highlighting the importance of considering environmental influences for robust performance evaluations.
What Problem Does This Paper Address?
The air transport sector is crucial for economic development, but its operation is resource-intensive. Accurately evaluating its efficiency is vital for sustainable growth and effective resource allocation. However, previous studies often overlooked the significant impact of external, uncontrollable factors (like regional GDP, technological development, and market openness) on operational performance. This gap makes it difficult to pinpoint true managerial inefficiencies versus those stemming from the operating environment. This paper aims to bridge this gap by providing a more nuanced evaluation of China's provincial air transport sector efficiency, disentangling managerial performance from environmental influences and statistical noise.
Methods
The core of this research is the application of a three-stage Data Envelopment Analysis (DEA). This method is an advancement over traditional DEA by explicitly addressing environmental factors and statistical noise.
- Stage 1: Initial Efficiency Evaluation: A standard DEA model (specifically, the Banker, Charnes, and Cooper (BCC) model, which accounts for variable returns to scale) is used to calculate the initial pure technical efficiency (PTE) and scale efficiency (SE) for each provincial air transport sector (DMU). This stage identifies overall inefficiencies, including those due to environmental factors and noise.
- Stage 2: Decomposing Inefficiencies: The input "slacks" (the difference between actual and ideal input levels) from Stage 1 are then regressed against a set of environmental variables using a Stochastic Frontier Analysis (SFA). SFA is crucial here as it allows for the decomposition of the observed slacks into three components: managerial inefficiency, environmental effects, and statistical noise. This stage quantifies the impact of each factor.
- Stage 3: "Real" Operational Efficiency: Using the insights from Stage 2, the input data is adjusted to remove the influence of environmental factors and statistical noise. A second DEA (BCC) analysis is then performed on these adjusted inputs. This final stage provides a "level playing field" assessment, reflecting the true operational performance attributable solely to managerial decisions and practices.
This three-stage approach is an extension of earlier four-stage DEA models, specifically incorporating SFA to better isolate managerial inefficiency from random noise.
Datasets
The study utilizes data for China's 30 provincial air transport sectors for the year 2017. The data sources include:
- China Statistical Yearbook: For capital stock and environmental variables.
- Statistical Yearbook of the Chinese Investment in Fixed Assets: For capital stock data.
- Year Book of China Transportation and Communications: For labor input data.
- Civil Aviation Administration of China (CAAC): For output data (passenger and freight throughput, aircraft movements).
The paper also constructs specific indicators:
- Capital Stock: Calculated using the perpetual inventory method.
- Infrastructure Construction Level (ICL): An index reflecting the number and level of airports based on ICAO standards.
The environmental variables used are:
- Gross Domestic Product (GDP) per Capita
- Actual Utilization of Foreign Direct Investment (FDI)
- Transaction Value in Technical Markets (TVTM)
- Three Kinds of Patents Granted per 10,000 People (TKPG)
- Household Consumption Expenditure (HCE)
Key Insights and Ideas
- Environmental Factors Matter: The study strongly demonstrates that regional socio-economic conditions significantly influence air transport operational efficiency. Ignoring these factors leads to a distorted view of managerial performance.
- Stage 1 vs. Stage 3 Differences: The efficiency scores in Stage 3 (after accounting for environmental factors and noise) show considerable differences from Stage 1. This highlights that some provinces appear less efficient not due to poor management, but because they operate in less favorable environments.
- "Real" Efficiency is Higher on Average: After adjusting for external factors, the average pure technical efficiency (PTE) increased from 0.894 to 0.916, suggesting that the initial assessment might have underestimated true managerial capabilities in some regions. Conversely, scale efficiency (SE) slightly decreased, indicating that environmental factors might have masked some scale inefficiencies.
- Regional Disparities are Pronounced: The analysis reveals significant differences in efficiency across China's regions, with North China and Northeast China showing distinct patterns in PTE and SE.
- Returns to Scale are Dynamic: The classification of returns to scale (IRS, CRS, DRS) changes significantly between Stage 1 and Stage 3 for many provinces, underscoring the impact of environmental adjustments on strategic recommendations.
Key Results
- Average Efficiency: In Stage 1, the average PTE was 0.894 and SE was 0.915. In Stage 3, after adjustments, the average PTE rose to 0.916, and the average SE dropped to 0.880.
- Number of Efficient Units: The number of purely technically efficient units (PTE=1) increased from 16 in Stage 1 to 18 in Stage 3.
- Regional Performance:
- North China and Northeast China showed the highest and lowest average PTE, respectively, in Stage 3.
- Central and Southern China had the highest scale efficiency, while North China had the lowest.
- North China and Northwest China, despite high PTE in Stage 3, showed lower SE, suggesting they operate at sub-optimal scales and could benefit from expansionary strategies.
- Environmental Impact: The SFA results (Table 4) indicated that GDP per capita and patent grants positively influenced input slacks (CS and ICL), suggesting that economic growth and technological advancement, while driving demand, can also increase operational challenges. Conversely, FDI utilization and household consumption expenditure had negative impacts on slacks, indicating they are beneficial for efficiency.
Limitations and Future Work
- Data Granularity: The study uses provincial-level data. Future research could benefit from more granular data (e.g., city-level) for a finer-grained analysis.
- Dynamic Analysis: This study is cross-sectional (2017 data). Incorporating panel data over time would allow for the analysis of dynamic efficiency changes and trends.
- Indicator Selection: While a comprehensive set of indicators was used, the selection of environmental variables could be expanded to include other relevant socio-economic or policy factors.
- Specific Policy Recommendations: While the paper provides general strategic implications (e.g., expansionary strategies for certain regions), more detailed, province-specific policy recommendations could be developed.
- DEA Model Extensions: Exploring other DEA variants, such as network DEA or meta-frontier DEA, could offer additional insights into the complex structure of the air transport sector.
Summary updated